Month: April 2019

Bundesgesetzblatt as Open Source

https://offenegesetze.de

Christian Endt: Aktivisten stellen alle Bundesgesetzblätter ins Netz, Süddeutsche Zeitung 10.12.2018

Each law contains a double fiction. Once it creates a conceivable state that is called linguistically, but must first be enforced by the threat and real enforcement of sanctions. On the other hand, the proclamation of the law contains the fiction that the law has already become known to the citizen. This fiction is called “irrefutable presumption” from a legal point of view, however, and instead of actual knowledge, the possibility of knowledge is sufficient. Otherwise law could not function at all as a self-referential system. It is the basis of the validity of the law to which the citizen has to adhere after the proclamation.
It is therefore all the more absurd to hide the promulgation gazettes behind a payment barrier – as happened with the Bundesgesetzblatt (German Federal Law Gazette) and also with some of the law and ordinance gazettes of the federal states of Germany. “Hiding” also means offering only a “read-only version” that cannot be printed out. One could even ask the question whether a proper proclamation has taken place.
The Act on Copyright and Related Rights (Copyright Act) (Gesetz über Urheberrecht und verwandte Schutzrechte (Urheberrechtsgesetz)) has taken this into account:
“§ 5 Official works: (1) Laws, ordinances, official decrees and announcements as well as decisions and officially drafted guidelines on decisions enjoy no copyright protection.”
However, the Bundesanzeiger-Verlag takes the view that its published laws, i.e. pdf files, whose contents are supplied by the state, enjoy database protection. However, this absurd assertion contains the threatening potential of a possible and uncertain court case, which many administrative lawyers would like to avoid in the interests of a smooth career.
The information specialist, however, has always been aware that a parliamentary or legal information system also includes free access to both promulgation bulletins and consolidated laws. The EU has therefore been making all public legal documents freely and electronically accessible for some time now. But the Parlamentsspiegel, a documentation system that wanted to make the parliamentary initiatives of the federal states – apart from the parliamentary databases there – available across the board, also scanned the Federal Law Gazette and the promulgation gazettes of the federal states and made them available online. However, this service was discontinued in 2014. It was not the information specialist, but the administrative lawyer as the objector who cut off the information for the citizen.
The company Makrolog has created a commercial product for Germany with its Recht für Deutschland service – also in accordance with the Copyright Act through its own digitalisation of the promulgation sheets.
The Open Knowledge Foundation Germany now offers all editions of the Federal Law Gazette freely accessible with OffeneGesetze.de. A full-text search complements the display by volumes and numbers. This grateful private initiative shows how miserable it still is in terms of digitisation with state initiative. Digitisation does not just start with artificial intelligence. Also in the matter of consolidated law, the private site www.buzer.de shows how conveniently, up-to-date and freely accessible one can present legal information: The text of the law, earlier versions with the possibility of creating synopses, reference to the official justification, citation in regulations, current changes, pending changes, search for regulations, law and full text, updating with web widget, feed, mail – it couldn’t be better. On the other hand, the Gesetze im Internet of the Federal Ministry of Justice and Consumer Protection, which are not the official version even there, and also dejure.org clearly fall off.
As far as the official version is concerned, the “Act supplementing Article 120 and amending Article 121 of the Constitution of the State of Hesse” : “The Journal of Laws and Decrees may be kept in electronic form in accordance with a law” adopted in the referendums on the amendment of the Constitution of Hesse on 28 October 2018 may be a step in the right direction.

Data as Menetekel 7: More than a quarter of the tweets on the migration pact Social Bots?

On 10.12.2018 the WELT published an article by Jan Lindenau: Robots mobilize against migration pact (behind the pay barrier). According to the article, Botswatch evaluated an analysis of 800,000 tweets between 24 November and 2 December. As a result, it announces that over a quarter of these tweets were social bots that created a mood against the migration pact in terms of content. This news was spread throughout the German press.
However, the analysis itself and the methodology proclaimed as a trade secret were not disclosed. This and the incomprehensible results met with considerable criticism from experts. Social media analyst Luca Hammer criticized the statements on Twitter. In a sample, he found that the proportion of bots – including automated information bots from the press – is around 6 percent. Data journalist Michael Kreil has written an open letter to Botswatch:
“We urgently need scientifically based, methodologically correct analyses of these processes! What we don’t need are actors who kidnap the discourse with unproven allegations, spread panic and uncertainty, and inadequately advise the federal government. For me, it is currently not possible to tell which position you would like to take with botswatch. Therefore I ask you, Mrs Wilke, to release the methods and data of your study “Social Bots and Migration Pact” for an independent, scientific review.”

The course of the discussion is described in the articles by Robert Tusch: Kritik an Botswatch: Warum die Debatte um die Social Bot-Studie zum Migrationspakt für Medien ist wichtig, in: Meedia vom 12.12.2018 sowie von Markus Reuter: Ein Bot allein macht keine Revolte. And also no migration debate, reproduced in: netzpolitik.org of 10.12.2018. It also deals with the methodological problems of identifying bots and other forms of accounts, which are more important for political influence on the net. “Alone in view of the influence of humanly operated accounts, it is simply dubious and subcomplex to try to explain social movements or hard-fought social discourses with bots.”
Jonas Hermann: Did bots have a one-sided influence on the debate on the migration pact? Neue Zürcher Zeitung of 13.12.2018 summarises the criticism and discussion once again. He points to Botswatch’s close personal ties to the CDU.
“The bot analysis was published on the day the migration pact was passed. This may or may not be a coincidence. What is certain, however, is that it can be used to discredit critics of the Pact and to present the debate about it as inflated and externally controlled.”
Thus, it is possible that it is not social bots, but freely invented data about social bots that should influence political discourse.

Data as Menetekel 6: Nitrogen dioxide limit invented

Alexander Kekulé, Director of the Institute for Medical Microbiology of the University Hospital Halle (Saale), has written an article in the ZEIT from 08.11.2018 – online behind the payment barrier – “Hysteria around the wrong thing. the limit of 40 micrograms for the exhaust gas is taken from the air”. To be read as facsimile on http://www.gegenwind-saarland.de/Klimawandel/181108-ZEIT-Hysterie-um-NO2.mrkd.pdf-Hysterie-um-NO2.mrkd.pdf . A shortened version was published in the Tagesspiegel of 10.11.2018, but no longer available on the online page of the TAGESSPIEGEL, but only via PRESSREADER. https://www.pressreader.com/germany/der-tagesspiegel/20181110/281715500636547
Subsequently, in 1993, the EU decided to set long-term air quality objectives, including a strict NO limit value. The legitimacy of such a value had to be based on the work of the World Health Organization (WHO). In order to reduce the then valid WHO guideline value of 150 micrograms, a working group was formed there, which made use of a meta-analysis from a five-year older report by the US Environmental Protection Agency. According to Kekulé, the meta-analysis evaluated various research projects whose parameters were completely different. Despite these completely different influencing factors, the result was “that respiratory diseases are 20 percent more frequent in households with gas cookers than in households with electric cookers”. However, no measured values could be derived from this, as very different concentrations (8 – 2500 micrograms) were found in households with gas cookers. “In the absence of useful data, the experts estimated without further ado that a gas stove increases the mean annual NO concentration in the household to about 40 micrograms and suggested this value as a guideline. To date, there is no evidence that the figure has anything to do with the health effects of NO”. Without verifying this, the EU adopted it as the legal limit. “To date, there are no robust data supporting the 40 microgram limit.”

Are statistical data facts?

In an article dated 02.11.2018 in the Neue Zürcher Zeitung, the article “The Federal Statistical Office creates the facts for Swiss democracy. Then it makes a mistake” how statistical data were officially changed:
“The statistics on the expulsion of foreign criminals showed that the state had allegedly expelled only 54 percent of them – the figure suggested that the others had been classified as hardship traps and had therefore not been expelled from the country. The topic is explosive, so far the effect of the expulsion initiative could not be quantified. After the first indignation, the BfS recalculated: 69 percent. Then it deleted the statistics completely from the website.”
Are data complex so that they need to be reduced in complexity in order to be comprehensible? Are statistical data facts or are they interpretations? What interpretations do you agree on or are the interpretations fake news?
“Of course statistics are highly political,” says Ulrich (the director). Because what is measured and how is political, who is asked and what questions are asked. “There is no truth, one can only agree on what one wants to measure. How high is poverty? Education? The gross domestic product? You make politics with indicators because you have agreed on them. “But in the end, everyone looks for their own truth in the statistics, that has always been the case,” says Ulrich.
“The withdrawal of BfS’s deportation figures shows that it is always necessary to agree again on what the facts are and how they are created. But the Fake News calls have sown one doubt: that there is such a thing as facts, that there is anything at all that has been agreed upon in this country.”

Facebook: the evil world of correlating data

Is Facebook to blame for the attacks on refugees in Germany?

The study by Karsten Müller and Carlo Schwarz: Fanning the Flames of Hate: Social Media and Hate Crime, Working Paper Series, University of Warwick No. 373 May 21, 2018 comes to the conclusion: “Our results suggest that social media can act as a propagation mechanism between online hate speech and real-life violent crime.”
First, the authors equate the use of the AFD (Alternative für Deutschland) Facebook page with the use of all right-wing social media pages. “In our setting, the share of a municipality’s population that use the AfD Facebook page is an intuitive proxy for right-wing social media use.” The general local use of local media is identified with the use of the “Nutella” page: “We thus attempt to isolate the local component of social media usage that is uncorrelated with right-wing ideology by drawing on the number of users on the “Nutella Germany” page”.
Jonny Häusler sums up the result on WIRED: “The study comes to the conclusion that there was a significant increase in violence against fugitives in small and large, liberal and conservative, poorer and richer communities if the average Facebook use in these communities was above the national average. In short, “Those who are too much on Facebook become violent against fugitives.”
So by using the “Nutella” Facebook page, you become a beast? Extremely unlikely, but we don’t want to ironize the efforts. It’s difficult enough to investigate the impact of the Internet and social media. But it remains so: correlations do not prove causality. Thus economics professor Tyler Cowen draws a sober balance on MARGINAL REVOLUTION: “As it stands right now, you shouldn’t be latching on to the reported results from this paper.”

Data as mood enhancer fake

In addition to its function as a warning sign, data also serves to brighten the mood:
Heiner Flassbeck: “About Fake News and Misappropriated Truths”, Makroskop 29.08.28, describes how the slightly increased IFO index (ifo Institute – Leibniz Institute for Economic Research at the University of Munich) in the last month was acclaimed in today’s news, but this does not correspond to the overall data situation: “The moderator’s jubilation and his statement are very close to a false report, because the mood is not “extremely good”, but … better than a month ago, but still far worse than at the turn of the year. Expectations have risen particularly, which may indeed have something to do with political news, while the assessment of the situation has changed only slightly.”
Mark Schieritz: “Five to eight / European Central Bank: The myth of the expropriation of savers” , in: ZEIT online 30.08.2018 claims that the financial assets of German savers have not suffered a loss due to low interest rates and inflation. This is a myth. (The real figures can be found here, for example: Interest rate gains and losses for German savers. Development of interest rate gains and losses from 2003 to 2018. tagesgeldvergleich.net) The reason he cites is that the financial assets of private households in Germany have increased overall since 2010 – for whatever reason. “One would like to be expropriated a little more”. The reason for the alleged myth of the loss of interest: “The myth is supposed to fuel resentment against the EU”. In any case, he tries to prevent this by using a data mood enhancer fake.